IN THE UNITED STATES, THERE is a high rate of HIV coinfection in persons with syphilis. In a review of 30 U.S. studies, median HIV seroprevalence rates were 27.5% in men with syphilis and 12.4% in women with syphilis, suggesting that persons with syphilis should be an important target group for HIV prevention.1 The importance of directing HIV prevention efforts toward persons with syphilis is underscored by studies that have demonstrated that syphilis facilitates the transmission and acquisition of HIV.2,3
Just as HIV infection is more prevalent in persons with syphilis than in those who do not have syphilis, HIV-infected persons may be more likely to acquire syphilis than those who are not infected with HIV. Several studies have demonstrated that persons infected with HIV have high rates of sexually transmitted diseases (STDs), although there is considerable variation in these rates across populations and geographic areas.4–9 Higher rates of STDs in HIV-infected persons might be attributable to behavioral factors, biologic factors (such as increased susceptibility to STDs resulting from immunosuppression), or both.3 In this study, we used data on the number of primary and secondary (P&S) syphilis cases reported in 2002 to estimate the rate of P&S syphilis in HIV-infected persons nationwide. These estimates can be used to help make informed decisions about the provision of STD and HIV prevention and care services by health departments, community agencies serving at-risk populations, and public and private medical care providers.
Methods
We estimated P&S syphilis rates in all women and men living with HIV in the United States in 2002. To estimate P&S syphilis rates in HIV-infected women, we approximated the number of new cases of P&S syphilis in HIV-infected women and divided by the estimated number of women living with HIV. The numerator (the number of new cases of P&S syphilis in HIV-infected women) was estimated by multiplying the number of reported P&S syphilis cases in women (CASESW) by the estimated percentage of these women who are also infected with HIV (COINFECTW). The denominator (the number of women living with HIV) was approximated by multiplying the estimated number of persons living with HIV (POPHIV) by the estimated percentage of these persons who are women (PCTW). Specifically, we calculated P&S syphilis rates in women with HIV as
;)
We estimated P&S syphilis rates in HIV-infected men in an analogous manner. The same equation was used except that values for the terms with subscript w were replaced with values specific to men. The estimated rates for men were divided into two categories: men who have sex with men (MSM) and men who have sex only with women (MSW). For simplicity, we assumed that the MSW category contained all men who were not in the MSM category.
Values for Equation Variables
We obtained base case values for the equation variables from the literature (Table 1). The reported number of P&S syphilis cases in women and men was obtained from national surveillance reports.10 The percentage of P&S syphilis cases in men that occurred in MSM (61.8%) was based on a recent modeling study.11
TABLE 1: Summary of Equation Variables Used to Estimate Primary and Secondary (P&S) Syphilis Rates in Persons With HIV in the United States, 2002
The estimated percentage of persons with syphilis who are also infected with HIV was based on HIV coinfection rates in persons with syphilis reported in three recent studies: a study of persons with syphilis in areas of persisting syphilis12 (HIV coinfection rate of 4% in persons with early syphilis), a study of unrecognized HIV infection in patients attending 28 STD clinics in 14 cities13 (HIV coinfection rate of 10% in persons with P&S syphilis), and a study of HIV seroincidence in persons attending 13 STD clinics in nine cities14 (HIV coinfection rate of 8% in persons with any stage of syphilis). We applied the average of these three estimates (7.3%) as the base case value for the percentage of MSW with syphilis that is coinfected with HIV. A review of studies of HIV prevalence in persons with syphilis indicated that the HIV coinfection rate in women with syphilis is roughly half that of men.1 Therefore, we used a base case coinfection rate of 3.6% for women, half the base case value for MSW. We used a base case coinfection rate of 46.5% for MSM with syphilis, based on coinfection rates ranging from 20% to 73% documented in syphilis outbreaks in MSM in several U.S. cities.15
We used a base case estimate of 925,000 for the number of people living with HIV in the United States in 2002 based on recent estimates derived in part from national surveillance data.16,17 We estimated that men accounted for 75% of the HIV-infected population, based on estimates that approximately 70% of new HIV infections occur in men and that approximately 80% of adults and adolescents living with AIDS in 1999 were men.18,19 We assumed that MSM accounted for 65% of all men with HIV in 2002 because the two HIV exposure categories of MSM and MSM who inject drugs accounted for approximately 65% of the new HIV cases reported in men from 1999–2002 (in states with HIV reporting) and approximately 65% of the adolescent and adult men living with AIDS in 1999.19,20
Sensitivity Analyses
We assigned a range of possible values to each variable except the reported number of P&S syphilis cases, which was available from surveillance reports. The ranges for four variables (HIV coinfection rates in MSM, number of people living with HIV in 2002, percentage of people living with HIV that are men, and the percentage of men living with HIV that are MSM) were selected so that their lower and upper bounds were consistent with ranges suggested by the sources from which base case values were obtained. We were unable to establish lower- and upperbound values for the three remaining variables (HIV coinfection rates in MSW, HIV coinfection rates in women, and the percentage of P&S syphilis cases in men that occurred in MSM) based on the sources from which their base case values were calculated. Therefore, the ranges for these variables were calculated as ±50% of the base case values to allow for a wide range of possible values.
After calculating the base case results, we performed two types of sensitivity analyses. First, we performed univariate sensitivity analyses to determine how the estimated P&S syphilis rates would change when we varied each variable individually from its lowerbound value to its upperbound value while holding other variables at their base case values. When we varied HIV coinfection rates, coinfection rates for women, MSW, and MSM were varied simultaneously. Second, we performed a multivariate analysis to examine how the estimated P&S syphilis rates would change when varying all variables simultaneously. Specifically, we conducted a Monte Carlo simulation21 in which we randomly assigned to each parameter a value between its lower- and upperbound estimates (assuming a uniform distribution) and then recalculated the estimated rates of P&S syphilis in persons with HIV. We performed this procedure 10,000 times to obtain a distribution of the estimated P&S syphilis rates.
Results
Under base case assumptions, the overall rate of P&S syphilis was estimated to be 186 new cases per 100,000 persons living with HIV in 2002 (Table 2). The rate of P&S syphilis was estimated to be 25 cases per 100,000 HIV-infected women, 60 cases per 100,000 HIV-infected MSW, and 336 cases per 100,000 HIV-infected MSM. Approximately one fourth of all P&S syphilis cases (1718 of 6862) in 2002 were estimated to have occurred in persons coinfected with HIV (Table 2).
TABLE 2: Estimation of Number of Primary and Secondary (P&S) Syphilis Cases, Number of P&S Syphilis Cases in Persons With HIV, Number of Persons Living With HIV, and P&S Syphilis Rates in Persons With HIV, United States, 2002
In the univariate sensitivity analyses, estimated overall rates of P&S syphilis ranged from 81 to 351 cases per 100,000 persons living with HIV (Table 3). Estimated rates of P&S syphilis ranged from 12 to 47 cases per 100,000 HIV-infected women, 12 to 114 cases per 100,000 HIV-infected MSW, and 144 to 634 cases per 100,000 HIV-infected MSM. In the multivariate sensitivity analyses, the estimated rates of P&S syphilis per 100,000 persons with HIV were between 70 and 391 overall, 11 and 49 for women, 13 and 147 for MSW, and 115 and 751 for MSM in 90% of the simulations (Table 3).
TABLE 3: Range of Estimated Primary and Secondary (P&S) Syphilis Rates (New Cases per 100,000 Persons) in Persons With HIV in the United States, 2002, when Applying Lower- and Upper Bounds of Equation Variables: Univariate and Multivariate Sensitivity Analyses
Discussion
The estimated rate of P&S syphilis in persons with HIV (186 per 100,000 persons) in 2002 was considerably higher than the reported rate of P&S syphilis in the general population of the United States that year (2.4 cases per 100,000 persons).10 Furthermore, the estimated rates of P&S syphilis in women, MSW, and MSM living with HIV (25, 60, and 336 per 100,000 persons, respectively) were higher than reported rates in the general population in the age groups with the highest rates of P&S syphilis in 2002 (3.3 cases per 100,000 women aged 20–24 years and 9.9 cases per 100,000 men aged 35–39 years).10 High rates of P&S syphilis in persons with HIV are a particularly troubling public health matter because syphilis is a marker for behaviors associated with HIV transmission,22 and the presence of infectious syphilis can facilitate the transmission and acquisition of HIV.2,3
The finding that the estimated P&S syphilis rates in persons with HIV are substantially higher than that of the general population has important implications for how public health programs develop and implement disease intervention activities. Changes in healthcare delivery and healthcare access behavior by populations at risk for STD and HIV may reduce opportunities for syphilis detection and prevention. For example, some public STD clinics have reduced services and hours, and emergency rooms that may provide care for persons at risk for syphilis rarely offer routine risk assessment and syphilis screening.23–27 Medical care providers in public and private sector settings, including HIV and infectious disease specialists, play an integral role in the continuing care of persons infected with HIV.28 Mathematical models suggest that treatment of STDs in persons receiving care for HIV could have a substantial impact on HIV transmission.29–31
Current guidelines for incorporating HIV prevention into the medical care of persons living with HIV suggest screening for syphilis at least once per year.28,32 More frequent screening is advised for persons with high-risk sexual behaviors or persons residing in communities with high syphilis incidence.28,32 The 2002 Centers for Disease Control and Prevention (CDC) Sexually Transmitted Disease Treatment Guidelines recommend that sexually active MSM receive a syphilis serology at least once a year (regardless of HIV status), although more frequent screening may be required for MSM engaged in high-risk behavior.33 Our results indicate that MSM living with HIV are at increased risk of acquiring syphilis, suggesting that more frequent screening of sexually active MSM with HIV may be warranted in areas that are experiencing, or are likely to experience, syphilis outbreaks in MSM.
Evidence suggests, however, that periodic sexual risk assessment and syphilis screening is not routinely provided to many HIV-infected persons in care.30,34 Interventions to improve access and quality of syphilis risk assessment, screening, and treatment are needed to promote relevant policies, protocols, and practices. Examples include promoting syphilis screening by tracking and evaluating screening rates and feeding back to the providers and reminder systems to encourage syphilis screening services.35,36
Our study is subject to several limitations. Although only three main variables were used in the analyses (the number of P&S syphilis cases, the number of people living with HIV, and the percentage of P&S syphilis cases that occur in persons with HIV), there is some uncertainty regarding the values for these variables. The number of P&S syphilis cases was based on reported cases and was not adjusted to account for the possibility that 20% to 50% of new syphilis cases are not reported.37,38 Adjusting for underreporting would have increased the estimated number of P&S syphilis cases (and therefore the estimated P&S syphilis rate) in HIV-infected persons. The percentage of P&S syphilis cases in men that occurred in MSM was based on a modeling study of the changes in the male-to-female ratio of reported cases and is subject to uncertainty.11 To address this uncertainty, we used a wide range of possible values of this variable in the sensitivity analysis.
Our results were sensitive to changes in the estimated number of people living with HIV. The estimated rates of P&S syphilis were approximately three times higher when applying the upperbound value of the number of people living with HIV than when applying the lowerbound value. Such a wide range of estimated rates of P&S syphilis when varying the number of people living with HIV would be expected, given that the upperbound value for the number of people living with HIV was approximately three times higher than the lowerbound value.
Changes in the HIV coinfection rates had a similar effect on our results. The estimated rates of P&S syphilis were approximately three to three and a half times higher when applying the upperbound value of HIV coinfection rates than when applying the lowerbound value. The range of possible P&S syphilis rates when varying HIV coinfection rates could be wider than we estimated if the ranges of values we assumed for HIV coinfection rates (±50% of the base case values) were too narrow. Even when applying the lowerbound values of the HIV coinfection rates, however, the estimated rates of P&S syphilis in persons with HIV were several times higher than those of the general population.
The base case values we applied for the HIV coinfection rates in persons with syphilis were based on studies of HIV prevalence over a range of years (1991–1997 for women and MSW and 1997–2001 for MSM), and these rates might not accurately reflect HIV coinfection rates among persons with syphilis in 2002. Furthermore, these studies reported HIV prevalence in persons with various stages of syphilis, not just P&S syphilis. Because some of the coinfected persons may have acquired syphilis before acquiring HIV, our application of these coinfection rates may have overestimated the percentage of people who were already HIV-infected at the time they acquired syphilis. Such overestimation would lead to an overestimation of P&S syphilis rates among persons with HIV to a comparable degree. In addition, coinfection rates in women and MSW were based primarily on persons attending STD clinics, whereas coinfection rates for MSM were based on men reported to have syphilis during recent syphilis outbreaks in urban areas. The base case coinfection rates estimated from these populations might not accurately reflect the national average coinfection rates of all persons who acquired syphilis in 2002.
Although the base case coinfection rates are subject to uncertainty and numerous potential biases, the base case coinfection rates we used were much lower than suggested in a comprehensive review of studies of HIV coinfection in persons with syphilis in the United States published from 1985 to 1998 (27.5%, 12.4%, and over 64% in men, women, and MSM with syphilis, respectively).1 Using the higher estimates of HIV coinfection rates from that study would have increased our estimated rate of P&S syphilis in HIV-infected persons substantially.
Our estimates of rates of P&S syphilis in HIV-infected persons are nationwide averages. In reality, the incidence of P&S syphilis in HIV-infected persons (as well as syphilis and HIV coinfection rates) vary substantially from one area to another and within a given area over time.1 For example, during a syphilis outbreak in King County, Washington, in MSM, the annual rate of P&S syphilis in HIV-infected MSM was projected to be 1500 per 100,000 in 1999,7 which is considerably higher than the national rate of P&S syphilis in HIV-infected MSM in 2002 that we estimated. Because syphilis rates in persons with HIV can vary across geographic areas, local surveillance data could be an important factor in determining the appropriate response to the potentially heightened risk of syphilis in HIV-infected persons in a given community.9
Despite limitations, this study provides useful estimates of national P&S syphilis rates in HIV-infected persons. The estimated rate of new P&S syphilis cases in HIV-infected persons is substantially higher than that of the general population. These findings highlight the importance of providing enhanced STD prevention services to HIV-infected persons.
References
1. Blocker ME, Levine WC, St. Louis ME. HIV prevalence in patients with syphilis, United States. Sex Transm Dis 2000; 27:53–59.
2. Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: The contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Transm Infect 1999; 75:3–17.
3. Wasserheit JN. Epidemiological synergy. Interrelationships between human immunodeficiency virus infection and other sexually transmitted diseases. Sex Transm Dis 1992; 19:61–77.
4. Scheer S, Chu PL, Klausner JD, Katz MH, Schwarcz SK. Effect of highly active antiretroviral therapy on diagnoses of sexually transmitted diseases in people with AIDS. Lancet 2001; 357:432–435.
5. Belongia EA, Danila RN, Angamuthu V, et al. A population-based study of sexually transmitted disease incidence and risk factors in human immunodeficiency virus-infected people. Sex Transm Dis 1997; 24:251–256.
6. Bersoff-Matcha SJ, Horgan MM, Fraser VJ, et al. Sexually transmitted disease acquisition among women infected with human immunodeficiency virus type 1. J Infect Dis 1998; 178:1174–1177.
7. Centers for Disease Control and Prevention. Resurgent bacterial sexually transmitted disease among men who have sex with men—King County, Washington, 1997–1999. MMWR Morb Mortal Wkly Rep 1999; 48:773–777.
8. Golden MR, Rompalo AM, Fantry L, et al. Early intervention for human immunodeficiency virus in Baltimore sexually transmitted disease clinics. Impact on gonorrhea incidence in patients infected with HIV. Sex Transm Dis 1996; 23:370–377.
9. Klausner JD, Stanley H, Stansell J. STD screening among HIV-infected patients in care, San Francisco. AIDS Patient Care STDS 2001; 15:73–76.
10. Sexually Transmitted Disease Surveillance, 2002. Atlanta: Department of Health and Human Services, Centers for Disease Control and Prevention, 2003.
11. Heffelfinger JD, Swint EB, Weinstock HS. Estimates of the number of cases of primary and secondary syphilis occurring among men who have sex with men in the United States, 1999–2002. Presented at the 2004 National STD Prevention Conference; 2004; Philadelphia.
12. Koumans EH, Farley TA, Gibson JJ, et al. Characteristics of persons with syphilis in areas of persisting syphilis in the United States: Sustained transmission associated with concurrent partnerships. Sex Transm Dis 2001; 28:497–503.
13. Weinstock H, Dale M, Linley L, et al. Unrecognized HIV infection among patients attending sexually transmitted disease clinics. Am J Public Health 2002; 92:280–283.
14. Weinstock H, Dale M, Gwinn M, et al. HIV seroincidence among patients at clinics for sexually transmitted diseases in nine cities in the United States. J Acquir Immun Defic Syndr 2002; 29:478–483.
15. Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 1999. MMWR Morb Mortal Wkly Rep 2001; 50:113–117.
16. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2004 Report on the Global HIV/AIDS Epidemic. Geneva: UNAIDS, 2004.
17. Fleming PL, Byers RH, Sweeney PA, et al. HIV prevalence in the United States, 2000. Presented at the 2002 9th Conference on Retroviruses and Opportunistic Infections; 2002; Seattle.
18. A Glance at the HIV Epidemic. Atlanta: Department of Health and Human Services, Centers for Disease Control and Prevention, 2001.
19. HIV/AIDS Surveillance Report 12[2]. Atlanta: Department of Health and Human Services, Centers for Disease Control and Prevention, 2001.
20. Centers for Disease Control and Prevention. Increases in HIV diagnoses—29 states, 1999–2002. MMWR Morb Mortal Wkly Rep 2003; 52:1145–1148.
21. Doubilet P, Begg CB, Weinstein MC, et al. Probabilistic sensitivity analysis using Monte Carlo simulation. A practical approach. Med Dec Making 1985; 5:157–177.
22. Erbelding EJ, Chung SE, Kamb ML, et al. New sexually transmitted diseases in HIV-infected patients: Markers for ongoing HIV transmission behavior. J Acquir Immun Defic Syndr 2003; 33:247–252.
23. Kimball AM, Lafferty WE, Kassler WJ, et al. The impact of health care market changes on local decision making and STD care: Experience in three counties. Am J Prev Med 1997; 13:75–84.
24. Centers for Disease Control and Prevention. Impact of closure of a sexually transmitted disease clinic on public health surveillance of sexually transmitted diseases—Washington, DC, 1995. MMWR Morb Mortal Wkly Rep 1998; 47:1067–1069.
25. Mehta SD, Rompalo A, Rothman RE, et al. Generalizability of STD screening in urban emergency departments: Comparison of results from inner city and urban sites in Baltimore, Maryland. Sex Transm Dis 2003; 30:143–148.
26. Mehta SD, Shahan J, Zenilman JM. Ambulatory STD management in an inner-city emergency department: Descriptive epidemiology, care utilization patterns, and patient perceptions of local public STD clinics. Sex Transm Dis 2000; 27:154–158.
27. Gunn RA, Rolfs RT, Greenspan JR, et al. The changing paradigm of sexually transmitted disease control in the era of managed health care. JAMA 1998; 279:680–684.
28. Centers for Disease Control and Prevention. Incorporating HIV prevention into the medical care of persons living with HIV: recommendations of CDC, the Health Resources and Services Administration, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. MMWR Recomm Rep 2003; 52:1–24.
29. Rothenberg RB, Wasserheit JN, St. Louis ME, et al. The effect of treating sexually transmitted diseases on the transmission of HIV in dually infected persons: A clinic-based estimate. Ad Hoc STD/HIV Transmission Group. Sex Transm Dis 2000; 27:411–416.
30. Farley TA, Cohen DA, Wu SY, et al. The value of screening for sexually transmitted diseases in an HIV clinic. J Acquir Immun Defic Syndr 2003; 33:642–648.
31. Chesson HW, Pinkerton SD. Sexually transmitted diseases and the increased risk for HIV transmission: Implications for cost-effectiveness analyses of sexually transmitted disease prevention interventions. J Acquir Immun Defic Syndr 2000; 24:48–56.
32. Centers for Disease Control and Prevention. HIV prevention through early detection and treatment of other sexually transmitted diseases—United States. Recommendations of the Advisory Committee for HIV and STD prevention. MMWR Recomm Rep 1998; 47:1–24.
33. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2002. MMWR Recomm Rep 2002; 51:1–78.
34. Rompalo A. The National Plan to Eliminate Syphilis From the United States: Recommendations for a Response to the HIV/AIDS Bureau. Report to the Health Resources and Services Administration, 2002.
35. New York State Department of Health AIDS Institute. The HIV Quality of Care Program. New York: New York State Department of Health, 2004.
36. Centers for Disease Control and Prevention. Congenital syphilis—United States, 2000. MMWR Morb Mortal Wkly Rep 2001; 50:573–577.
37. Hook EW III, Marra CM. Acquired syphilis in adults. N Engl J Med 1992; 326:1060–1069.
38. Cates W Jr. Estimates of the incidence and prevalence of sexually transmitted diseases in the United States. American Social Health Association Panel. Sex Transm Dis 1999; 26:S2–S7.
39. Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 2002. MMWR Morb Mortal Wkly Rep 2003; 52:1117–1120.
